llvm-project/clang/test/CodeGenCUDA
Sameer Sahasrabuddhe d8f99bb6e0 [AMDGPU] replace hostcall module flag with function attribute
The module flag to indicate use of hostcall is insufficient to catch
all cases where hostcall might be in use by a kernel. This is now
replaced by a function attribute that gets propagated to top-level
kernel functions via their respective call-graph.

If the attribute "amdgpu-no-hostcall-ptr" is absent on a kernel, the
default behaviour is to emit kernel metadata indicating that the
kernel uses the hostcall buffer pointer passed as an implicit
argument.

The attribute may be placed explicitly by the user, or inferred by the
AMDGPU attributor by examining the call-graph. The attribute is
inferred only if the function is not being sanitized, and the
implictarg_ptr does not result in a load of any byte in the hostcall
pointer argument.

Reviewed By: jdoerfert, arsenm, kpyzhov

Differential Revision: https://reviews.llvm.org/D119216
2022-02-11 22:51:56 +05:30
..
Inputs [HIP] Fix linking of asanrt.bc 2021-09-27 13:25:46 -04:00
address-spaces.cu [Clang/Test]: Rename enable_noundef_analysis to disable-noundef-analysis and turn it off by default 2022-01-16 18:54:17 +09:00
alias.cu
amdgpu-alias-undef-symbols.cu [HIP] [AlwaysInliner] Disable AlwaysInliner to eliminate undefined symbols 2021-10-18 16:53:15 -06:00
amdgpu-asan.cu [AMDGPU] replace hostcall module flag with function attribute 2022-02-11 22:51:56 +05:30
amdgpu-atomic-ops.cu Let clang atomic builtins fetch add/sub support floating point types 2021-04-06 15:44:00 -04:00
amdgpu-code-object-version.cu [HIP] Emit amdgpu_code_object_version module flag 2022-02-08 21:58:40 -05:00
amdgpu-func-attrs.cu [test] Add {{.*}} to make ELF tests immune to dso_local/dso_preemptable/(none) differences 2020-12-31 00:27:11 -08:00
amdgpu-kernel-arg-pointer-type.cu [Clang/Test]: Rename enable_noundef_analysis to disable-noundef-analysis and turn it off by default 2022-01-16 18:54:17 +09:00
amdgpu-kernel-attrs.cu Relands "[HIP] Change default --gpu-max-threads-per-block value to 1024" 2021-02-12 10:53:59 -05:00
amdgpu-visibility.cu [test] Add {{.*}} to make ELF tests immune to dso_local/dso_preemptable/(none) differences 2020-12-31 00:27:11 -08:00
amdgpu-workgroup-size.cu
atomic-ops.cu [HIP] Add atomic load, atomic store and atomic cmpxchng_weak builtin support in HIP-clang 2021-11-29 12:07:13 -07:00
atomics-remarks-gfx90a.cu [Remarks] Emit optimization remarks for atomics generating CAS loop 2021-08-16 14:56:01 -06:00
builtin-mangled-name.cu [CUDA][HIP] add __builtin_get_device_side_mangled_name 2021-03-25 15:25:29 -04:00
builtins-amdgcn.cu [Clang/Test]: Rename enable_noundef_analysis to disable-noundef-analysis and turn it off by default 2022-01-16 18:54:17 +09:00
builtins-unsafe-atomics-gfx90a.cu [HIP] Relax conditions for address space cast in builtin args 2021-10-15 15:35:52 -06:00
const-var.cu [CUDA][HIP] Do not treat host var address as constant in device compilation 2022-01-28 16:04:52 -05:00
constexpr-variables.cu
convergent.cu [Clang] [NFC] fix CHECK lines for convergent attribute tests 2021-06-29 00:21:07 +05:30
correctly-rounded-div.cu [HIP] Fix device lib selection 2021-05-01 20:31:11 -04:00
cuda-builtin-vars.cu [Clang/Test]: Rename enable_noundef_analysis to disable-noundef-analysis and turn it off by default 2022-01-16 18:54:17 +09:00
cxx-call-kernel.cpp Fix test cxx-call-kernel.cpp 2021-03-01 17:10:53 -05:00
debug-info-address-class.cu
debug-info-template.cu
deferred-diag.cu
dependent-libs.cu
device-fun-linkage.cu [test] Add {{.*}} to make ELF tests immune to dso_local/dso_preemptable/(none) differences 2020-12-31 00:27:11 -08:00
device-init-fun.cu
device-stub.cu [HIP] Fix -fgpu-rdc for Windows 2021-12-06 16:42:23 -05:00
device-use-host-var.cu CUDA/HIP: Change device-use-host-var.cu's NOT "external" check to include variable name 2021-06-04 13:10:00 -04:00
device-var-init.cu [test] Add {{.*}} to make ELF tests immune to dso_local/dso_preemptable/(none) differences 2020-12-31 00:27:11 -08:00
device-var-linkage.cu [CUDA] Fix static device variables with -fgpu-rdc 2021-08-25 09:31:22 +02:00
device-vtable.cu [test] Add {{.*}} to make ELF tests immune to dso_local/dso_preemptable/(none) differences 2020-12-31 00:27:11 -08:00
dft-func-attr-skip-intrinsic.hip [Clang] [NFC] fix CHECK lines for convergent attribute tests 2021-06-29 00:21:07 +05:30
filter-decl.cu
flush-denormals.cu [CUDA][HIP] rename -fcuda-flush-denormals-to-zero 2021-04-05 00:13:51 -04:00
fp-contract.cu [HIP] Fix regressions due to fp contract change 2020-11-24 08:10:06 -05:00
function-overload.cu [test] Add {{.*}} to make ELF tests immune to dso_local/dso_preemptable/(none) differences 2020-12-31 00:27:11 -08:00
host-used-device-var.cu [Internalize] Preserve variables externally initialized. 2021-07-08 10:48:19 -04:00
kernel-amdgcn.cu Relands "[HIP] Change default --gpu-max-threads-per-block value to 1024" 2021-02-12 10:53:59 -05:00
kernel-args-alignment.cu [Clang/Test]: Rename enable_noundef_analysis to disable-noundef-analysis and turn it off by default 2022-01-16 18:54:17 +09:00
kernel-args.cu [Clang/Test]: Rename enable_noundef_analysis to disable-noundef-analysis and turn it off by default 2022-01-16 18:54:17 +09:00
kernel-call.cu
kernel-dbg-info.cu [HIP] Emit kernel symbol 2021-03-01 16:31:40 -05:00
kernel-stub-name.cu [Clang/Test]: Rename enable_noundef_analysis to disable-noundef-analysis and turn it off by default 2022-01-16 18:54:17 +09:00
lambda-reference-var.cu [test] Make ELF tests immune to dso_local/dso_preemptable/(none) differences 2020-12-31 13:59:44 -08:00
lambda.cu [Clang/Test]: Rename enable_noundef_analysis to disable-noundef-analysis and turn it off by default 2022-01-16 18:54:17 +09:00
launch-bounds.cu
library-builtin.cu [test] Add {{.*}} to make ELF tests immune to dso_local/dso_preemptable/(none) differences 2020-12-31 00:27:11 -08:00
link-device-bitcode.cu [test] Add {{.*}} to make ELF tests immune to dso_local/dso_preemptable/(none) differences 2020-12-31 00:27:11 -08:00
llvm-used.cu Change some addUsedGlobal to addUsedOrCompilerUsedGlobal 2021-02-26 10:42:07 -08:00
managed-var.cu [CUDA] Fix static device variables with -fgpu-rdc 2021-08-25 09:31:22 +02:00
ms-linker-options.cu Revert "[hip][cuda] Enable extended lambda support on Windows." 2021-02-04 07:10:46 -05:00
norecurse.cu
nothrow.cu [test] Add {{.*}} to make ELF tests immune to dso_local/dso_preemptable/(none) differences 2020-12-31 00:27:11 -08:00
openmp-target.cu
printf-aggregate.cu
printf.cu
profile-coverage-mapping.cu [PGO][CUDA][HIP] Skip generating profile on the device stub and wrong-side functions. 2020-08-10 11:01:46 -04:00
propagate-metadata.cu [clang][cli] Remove -f[no-]trapping-math from -cc1 command line 2021-01-12 10:00:23 +01:00
ptx-kernels.cu [test] Add {{.*}} to make ELF tests immune to dso_local/dso_preemptable/(none) differences 2020-12-31 00:27:11 -08:00
redux-builtins.cu [Clang/Test]: Rename enable_noundef_analysis to disable-noundef-analysis and turn it off by default 2022-01-16 18:54:17 +09:00
static-device-var-no-rdc.cu [CUDA][HIP] Fix device template variables 2021-05-12 11:13:29 -04:00
static-device-var-rdc.cu [CUDA] Fix static device variables with -fgpu-rdc 2021-08-25 09:31:22 +02:00
surface.cu [Clang/Test]: Rename enable_noundef_analysis to disable-noundef-analysis and turn it off by default 2022-01-16 18:54:17 +09:00
texture.cu [Clang/Test]: Rename enable_noundef_analysis to disable-noundef-analysis and turn it off by default 2022-01-16 18:54:17 +09:00
types.cu [test] Add {{.*}} to make ELF tests immune to dso_local/dso_preemptable/(none) differences 2020-12-31 00:27:11 -08:00
unnamed-types.cu [Clang/Test]: Rename enable_noundef_analysis to disable-noundef-analysis and turn it off by default 2022-01-16 18:54:17 +09:00
unused-global-var.cu [Internalize] Preserve variables externally initialized. 2021-07-08 10:48:19 -04:00
usual-deallocators.cu [Clang/Test]: Rename enable_noundef_analysis to disable-noundef-analysis and turn it off by default 2022-01-16 18:54:17 +09:00
vtbl.cu [IRGen] Do not overwrite existing attributes in CGCall. 2022-01-20 13:45:19 +00:00